基于KPCA-WOA-KELM的岩爆烈度预测  

Classified Prediction Model of Rockburst Using KPCA-WOA-KELM

作　　者：郭延华[1] 赵帅 GUO Yanhua;ZHAO Shuai(School of Civil Engineering,Hebei University of Engineering,Handan,Hebei 056038,China)

机构地区：[1]河北工程大学土木工程学院,河北邯郸056038

出　　处：《河北工程大学学报（自然科学版）》2021年第2期1-7,共7页Journal of Hebei University of Engineering:Natural Science Edition

基　　金：河北省自然科学基金资助项目(E2014402099)。

摘　　要：岩爆是隧道开挖中常见的工程地质灾害,为准确预测岩爆烈度,提出基于KPCA-WOA-KELM的岩爆烈度预测模型。首先,根据岩爆烈度影响因素确定岩爆评判指标,并采用核主成分分析(KPCA)对岩爆数据做特征提取,简化模型输入参数的同时充分保留数据特征信息;其次,使用核极限学习机(KELM)拟合评判指标与岩爆烈度间的非线性映射关系,并采用鲸鱼优化算法(WOA)优化KELM的参数,避免人工设置参数对模型预测效果的影响;然后,使用准确率、精确率、召回率、F值等指标综合评估模型的预测性能;最后,利用秦岭终南山公路隧道岩爆实例验证该模型的可行性。研究表明,KPCA-WOA-KELM能有效地简化数据结构,避免局部最优解,提高岩爆烈度预测的准确率。Rockburst is a common engineering geological disaster in deep rock excavation.In order to predict rockburst intensity grade accurately,this paper proposes a rockburst intensity prediction model based on KPCA-WOA-KELM.Firstly,rockburst evaluation indexes are determined according to the influencing factors of rockburst intensity,and the kernel principal component analysis(KPCA)is used to perform feature compression on rockburst data,so as to simplify the input data structure of the model and fully retain the data feature information.Secondly,the kernel-based extreme learning machine(KELM)was used to fit the nonlinear mapping relationship between the evaluation index and rockburst intensity,and the whale optimization algorithm(WOA)is used to optimize the parameters of KELM to reduce the impact of manual setting parameters on the model prediction effect.Then,the accuracy,precision,recall,F-measure and other indicators are used to evaluate the prediction performance of the model.Finally,the prediction of rock burst intensity of Zhongnanshan highway tunnel in Qinling Mountains is made to verify the feasibility and applicability of the model.The results show that KPCA-WOA-KELM can simplify the data structure more effectively,effectively avoid the local optimal solution,and improve the accuracy of rockburst intensity prediction.

关 键 词：岩爆 核主成分分析 鲸鱼优化算法 核极限学习机 

分 类 号：TU45[建筑科学—岩土工程]